Instead of implementing MSW-specific code to handle HDC for GDI+ context
directly in wxGCDC delegate acquiring/releasing HDC to underlying
wxGraphicsContext. Decoupling GDI+-specific code from wxGCDC will allow
us to implement handling HDC in other graphics renderers in a clean way.
Currently it's only used for wxiOS build and the tests don't build
there.
Also set working-directory for the test building step once instead of
using "make -C" option twice.
Enable ellipsization to show the items reasonably well even if they're
too long to fit into the available space -- without this, just the tail
of the long items was shown and shorter items could have been not shown
at all, as only the blank part of their label was visible.
Closes https://github.com/wxWidgets/wxWidgets/pull/2403
Co-Authored-By: Kent Tessman <kent@generalcoffee.com>
Co-Authored-By: Paul Cornett <paulcor@users.noreply.github.com>
Co-Authored-By: Vadim Zeitlin <vadim@wxwidgets.org>
Allow using wxPrintf() etc with std::string_view arguments when using
C++17.
This is inefficient but still more convenient than not being able to use
string_view at all.
Define these functions in the class declaration to avoid gcc warning
about "redeclaring them without dllimport attribute after being
referenced with dll linkage".
This commit just moves the existing code around, no real changes.
Closes https://github.com/wxWidgets/wxWidgets/pull/2400
They were incorrectly interpreted as mnemonics when drawing
wxCheckListBox items, which didn't make sense and was inconsistent with
the other ports and even wxListBox in wxMSW itself.
It also affected wxRearrangeCtrl under MSW, which uses wxCheckListBox
for its implementation.
Closes#19201.
Don't reset the labels windows size to the default value, it should stay
as 0 because otherwise we break the invariants
m_rowLabelWidth != 0 <=> m_rowLabelWin->IsShown()
m_colLabelHeight != 0 <=> m_colLabelWin->IsShown()
that the rest of the code relies on.
Closes#18904.
Regenerate the files after the changes of 232d1c1156 (Feed sorted output
to xargs when generating wxstd.pot, 2021-01-23) which added sorting.
This results in tons of changes, but had to be done sooner or later and
now seems like not the worst moment to do it.
wxOVERRIDE was already used for the native MSW version, but not the
other ones, which prevented this header from compiling cleanly with
-Wsuggest-override and so broke allheaders test in these ports.
Closes https://github.com/wxWidgets/wxWidgets/pull/2396
This is required to avoid warnings (and worse, probably) due to
conflicting function declarations in the platform SDK and Cygwin headers
that have started happening since 8dfcb582da (Regenerate configure after
enabling IPv6 support by default, 2021-06-15).
This extends the changes of e0005c1d93 (No longer include the system
headers first, 2021-04-22) to the platform-specific headers included for
Cygwin/MinGW wxMSW builds, as this shouldn't be necessary neither any
more (but doing it for wxQt is still needed, for the reasons explained
in the new comment added by this commit).
It also incidentally fixes a problem in Cygwin build since enabling IPv6
in 8dfcb582da (Regenerate configure after enabling IPv6 support by
default, 2021-06-15) as including unistd.h directly results in warnings
if wx/defs.h, which predefines __USE_W32_SOCKETS, hadn't been included
before.
Actually test using the coreutils catalog, as this was broken for many
years due to using a wrong catalog name and not adding /usr/share/locale
as a search prefix, but went unnoticed.
So add a menu item to show translation of a string in this catalog to
make testing that this works simpler.
This is more clear than testing for Linux and will allow enabling this
code under other systems later (as there doesn't seem to be any real
reason not to do it under other Unix systems too).
No real changes.
There is no need to suppress wxLog messages from AddCatalog() because it
doesn't log any errors even if the catalog is not found, while tracing
messages (given only if WXTRACE is set) from it can be useful.
What used to be called "fileutils" is called "coreutils" since many
years now, so use it to demonstrate loading of the standard catalog.
Closes https://github.com/wxWidgets/wxWidgets/pull/2395
This is another library that can be detected and used by libtiff
configure that we don't want to use, as built-in libraries shouldn't
have any external dependencies, so explicitly disable its use, just as
it was already done for libjbig, libwebp and libzstd, see ec4b71afa0
(Update tiff submodule, 2018-11-17) and b7600b0b1b (Ensure jbig support
is disabled for builtin libtiff, 2020-06-17).
Closes https://github.com/wxWidgets/wxWidgets/pull/2394
Consistently check that the index is valid in all ports, instead of
using hard-coded 255 rather than the actual number of colours in some of
them and forgetting to check that the index is positive in others.
Closes#19198.
This shouldn't really change anything, but is more appropriate and might
matter in the future if other event handlers are added to wxBookCtrlBase
event table.
This was broken in to 682cb8355c (Replace "sendEvent" parameter in
wxGridSelection with "eventType", 2020-08-21) which changed Select() to
take wxEventType argument but still passed it "true" in a few places.
Fix this and add a unit test verifying that this works as expected.
Ideal would, of course, be to avoid implicit conversions from bool to
wxEventType but making wxEventType anything other than int, for which
such conversions are unavoidable, is too backwards-incompatible to
seriously consider.
the Move handler apparently misses some resize events, so move the
gst_player_video_overlay_video_renderer_set_render_rectangle call
to expose_event_callback.
This is kept as a separate commit because it would be more efficient
to keep it in Move once we can catch that initial size change, so
this commit can get reverted then.
gstreamer is known to crash on xvimagesink if the main window is
wayland-native and DISPLAY is set: try to make it not load.
Also do the same for ximagesink just in case.
gstreamer creates a new connection to the wayland display by default, and
gst_video_overlay_set_window_handle() only works if both the parent surface
(part of the gtk window) and the gstreamer surface are on the same display,
so we need to tell gstreamer about our wl_display when it asks
